Quick viewing(Text Mode)

Bossuyt2009chap49.Pdf

Bossuyt2009chap49.Pdf

and (Anura)

Franky Bossuyt and Kim Roelants of molecular phylogenetic studies is now leading to an Biology Department, Unit of Ecology & Systematics, increasingly resolved consensus for the anuran tree. In Lab, Vrije Universiteit Brussel, Pleinlaan 2, 1050 Brussels, this chapter, we review the relationships and divergence Belgium times among 59 anuran families and argue that their *To whom correspondence should be addressed (fbossuyt@vub. evolutionary history is largely congruent with major ac.be) geological and environmental changes in Earth’s history. We mostly implement and family names derived Abstract from the recently proposed by Frost et al. (3). However, we believe that evolutionary time is an import- Anura (frogs and toads) constitute over 90% of living parameter in conveying useful comparative infor- amphibian diversity. Recent timetree constructions have mation in biological classiA cation (4). We therefore treat shown that their diversifi cation was a highly episodic pro- Ascaphidae, Discoglossidae, Nasikabatrachidae, some cess, with establishment of the major in three peri- subfamilies in Nobleobatrachia, and all subfamilies of ods: (251–200 million years ago, Ma), end of sensu lato as distinct families. to early (~150–100 Ma), and end of Cretaceous 7 e sequence of early divergences in Anura has been to early Paleogene (~70–50 Ma). The early diversifi cation the subject of major controversy. Most of the debate of anurans predated the initial north–south breakup of focused on the phylogenetic position of archaeobatra- , and resulted in distinct assemblages in both hemi- chian families (taxa with primitive or transitional char- spheres. The subsequent radiation of neobatrachian frogs acters, covering ~4% of all extant ) with respect has been largely determined by Gondwanan fragmenta- to neobatrachian families (“advanced” taxa, covering tion and resulted in recurrent patterns of continent-scale endemism.

Anura (Fig. 1) (“tail-less ”) represent the lar- gest living of amphibians, and currently include ~5400 described species (1). Most of them undergo the typical amphibious life history and are dependent on the presence of water for their reproduction and devel- opment. Multiple lineages, however, show an evolution- ary trend toward increased terrestriality in larval or adult stages. Despite an evolutionarily conserved body plan, anurans have diversiA ed into a myriad of eco- morphs and have adapted to life in habitats as distinct as rainforest canopies, mangroves, and sand dune burrows. In addition, anurans have attained a subcosmopolitan distribution and are currently only absent in extreme northern latitudes, , and most oceanic (non- continental) islands (2). 7 e independent occupation of similar ecological niches by frog taxa in diB erent geo- graphic regions has resulted in extraordinary cases of evolutionary convergence. 7 e consequent high levels of morphological homoplasy have complicated anuran sys- Fig. 1 A (Rhacophorus lateralis) from India. Credit: tematics for decades. However, a major ongoing upsurge F. Bossuyt.

F. Bossuyt and K. Roelants. Frogs and toads (Anura). Pp. 357–364 in e Timetree of Life, S. B. Hedges and S. Kumar, Eds. (Oxford University Press, 2009).

HHedges.indbedges.indb 357357 11/28/2009/28/2009 11:28:26:28:26 PPMM 358 THE TIMETREE OF LIFE

Microhylidae 43 Dyscophidae 37 Asterophryidae 35 Kalophrynidae 46 Cophylidae 44 32 Melanobatrachidae

Gastrophrynidae Microhyloidea 31 Scaphiophrynidae 28 Hoplophrynidae 25 Phrynomeridae

17 42 26 Astylosternidae 21

38 Afrobatrachia Hemisotidae

41 36 16 33 39 Ranidae 29 Micrixalidae 40 34 24 Natatanura 11 23 30 27 Phrynobatrachidae 22 Nasikabatrachidae

Triassic Jurassic Cretaceous Pg Ng Sooglossoidea MESOZOIC CENOZOIC 250 200 150 100 50 0 Million years ago

Fig. 2 Continues

the remaining 96% of extant species). Morphological () as the closest relatives of Neo- studies have supported diverse paraphyletic arrange- batrachia (11–13). Recent phylogenetic studies, imple- ments of archaeobatrachian families (5–10). Although menting expanded taxon sampling, nuclear and Neobatrachia have traditionally been considered to con- mitochondria l protein-coding DNA sequences, and mod- stitute a single-nested clade, analyses of combined lar- el-based reconstruction methods, have provided robust val and adult characters have recently questioned their support for a paraphyletic arrangement of four major monophyly (9, 10). archaeobatrachian lineages: (i) Amphicoela: Ascaphidae Early analyses of ribosomal DNA sequences clus- + Leiopelmatidae, (ii) Costata (previously known as tered the archaeobatrachian families in a single clade Discoglossoidea), (iii) Xenoanura (Pipoidea), and (iv)

HHedges.indbedges.indb 358358 11/28/2009/28/2009 11:28:30:28:30 PPMM Eukaryota; Metazoa; Vertebrata; Lissamphibia; Anura 359

(continued from previous page) 56 52 10 55 51 49 54 Telmatobiidae 47 Nobleobatrachia Bufonidae 53 Dendrobatidae 9 45 48 50 Centrolenidae

13 Neobatrachia 20 Rheobatrachidae 19 4 15

Calyptocephalellidae Myobatrachoidea Heleophrynidae Pelobatidae 3 18 14 8

Scaphiopodidae Anomocoela 2 7

Alytidae Xenoanura 12 1 6 Discoglossidae

Bombinatoridae Costata Ascaphidae 5 Leiopelmatidae Amphicoela Triassic Jurassic Cretaceous Pg Ng MESOZOIC CENOZOIC 250 200 150 100 50 0 Million years ago

Fig. 2 A timetree of frogs and toads (Anura). Divergence times are shown in Table 1 Ng (Neogene), Pg (Paleogene), and Tr (Triassic)..

Anomocoela () (3, 14–18). Consistent with (14), of an Anomocoela + Neobatrachia clade (16–18), most morphological evidence, they supported the basal and of a Costata + Anomocoela + Neobatrachia clade divergence of Amphicoela, and identiA ed Anomocoela as (3, 15). the closest relative of Neobatrachia. A remaining point 7 e monophyly of Neobatrachia has always received of ambiguity is the phylogenetic position of Xenoanura. strong support from molecular data ( 3, 12–22). Vari- Molecular studies have variously resolved Xenoanura as ous arrangements of the following four well-supported the closest relative of Costata (17, 19), of Neobatrachia lineages have been published: (i) Heleophrynidae,

HHedges.indbedges.indb 359359 11/28/2009/28/2009 11:28:31:28:31 PPMM 360 THE TIMETREE OF LIFE

Table 1. Divergence times (Ma) and their 95% confi dence/credibility intervals (CI) among anurans.

Timetree Estimates Node Time Ref. (15) Ref. (18) Ref. (26) Ref. (30) Ref. (37) (Fig. 2) (Ma) Time CI Time CI Time Time CI Time

1 243 262 304–223 243 264–217 – – – – 2 234 245 288–204 234 258–210 183 – – – 3229––229253–204–––– 4 222 216 260–176 222 247–197 – – – – 5 203 237 281–193 203 227–177 172 – – – 6 203 199 244–155 203 228–180 132 – – – 7193––193218–170–––– 8 167 164 208–121 167 191–146 127 – – – 9 167 162 199–128 167 186–149 125 – – – 10 161 150 186–117 161 180–143 – – – – 11 159 131 167–99 159 178–141 110 – – – 12 152 152 199–105 152 179–129 – – – – 13 146 138 172–108 146 164–129 106 – – – 14 143 142 187–101 143 166–121 112 – – – 15 129 120 154–91 129 147–113 94.4 – – – 16 119 99 132–70 119 131–106 – 133 154–115 – 17 117 – – 117 129–104 – 127 148–110 – 18 115 118 162–77 115 136–93 93.3 – – – 19107––107123–89–––– 20102––102118–85–––– 21 102 – – 102 115–90 – 107 126–89 – 22101––101121–83–––– 23 94.4 – – 82.5 95–71 – 94.4 113–78 – 2489.7–––––89.7108–74– 25 88 – – 75.6 83–70 – 88 102–77 – 26 87.7 – – 87.7 100–75 – 96 117–78 – 2786.3–––––86.3104–71– 2883.9–––––83.997–74– 2982.9–––––82.9100–68– 30 81.1 – – 72.6 85–62 – 81.1 99–66 – 3180.5–––––80.593–72– 3279–––––7991–70– 3378.1–––––78.195–64– 3477.6–––––77.695–63– 35 76.8 – – 73.0 79–68 – 76.8 89–69 – 36 74.4 – – 70.3 81–60 – 74.4 91–61 – 37 73.6 – – 70.2 76–66 – 73.6 85–67 – 38 72.1 – – 72.1 85–59 – 74.6 94–58 – 3971.7–––––71.788–58– 40 71.2 – – – – – 71.2 89–56 –

HHedges.indbedges.indb 360360 11/28/2009/28/2009 11:28:31:28:31 PPMM Eukaryota; Metazoa; Vertebrata; Lissamphibia; Anura 361

Table 1. Continued

Timetree Estimates Node Time Ref. (15) Ref. (18) Ref. (26) Ref. (30) Ref. (37) (Fig. 2) (Ma) Time CI Time CI Time Time CI Time

41 69.7 – – 62.8 74–53 – 69.7 86–56 – 42 69.1 – – 69.1 81–57 – – – – 43 68.7 – – 67.1 71–66 – 68.7 78–65 – 4466.8–––––66.880–56– 4563––6377–53–––– 4661.8–––––61.876–49– 4761.1––61.176–51–––– 4859.7––59.774–50–––– 4958.4––58.472–49–––– 5057.9––57.972–49–––– 5156.7––56.770–47–––– 5255––5568–46–––– 5354.6––54.668–45–––– 5453.2––53.266–44–––– 5546.5–––––––46.5 56 43.2 – – 43.2 55–36 – – – –

Note: Node times in the timetree are based on refs. (18) and (30) for Natatanura and Microhyloidea, because the use of time estimates averaged across all studies would be incompatible with the depicted topology.

(ii) Sooglossoidea: Nasikabatrachidae + Sooglossidae, (= Caudiverbera) and (iii) Nobleobatrachia (the “” of 14, 15, 21) + Myo- (both now ) of Chile as the closest batrachoidea + Calyptocephalellidae, and (iv) Ranoides: relatives of the Australo-Papuan Myobatrachoidea (3, 15, Afrobatrachia + Microhyloidea + Natatanura. Most 18, 24). 7 e same analyses recovered this previously studies of the past few years have converged on a basal unrecognized clade (Australobatrachia) as the clos- split between the South African endemic Heleophry- est relatives of Nobleobatrachia. Several controversies nidae and the remaining neobatrachians (3, 14–18). An however remain in nobleobatrachian phylogeny: A rst, important addition to the amphibian tree resulted from Leptodactylidae as deA ned here was not found mono- the discovery of a new frog lineage (Nasikabatrachidae) phyletic in a recent molecular study (25), although this in the Western Ghats of India (21). All molecular stud- clade is supported by a unique insertion of two codons in ies have found this lineage to be the closest relative of the Ncx1 gene (18). Second, were vari- Sooglossidae, a small family endemic to the Seychelles ously found to be monophyletic (26) or polyphyletic (3). (3, 18, 21, 22). Most important, the sequence of rapid diversiA cation in Despite the absence of derived morphological char- the radiation of nobleobatrachian families can be consid- acters supporting Nobleobatrachia, its monophyly is ered largely unresolved. Ranoides are composed of three strongly supported by DNA sequence evidence of dif- highly supported family assemblages: Afrobatrachia, ferent loci, and includes a unique codon insertion in Microhyloidea, and Natatanura. Afrobatrachia repre- the RAG-1 gene. Within this clade, Leptodactylidae and sents a well-resolved African endemic clade and has Hylidae as traditionally deA ned (1, 2) are now known to recently been shown to include the Brevicipitidae ( 27), be polyphyletic (3, 18, 23, 24) and several of their lin- which were long considered part of the microhyloid eages have been assigned to separate families (3). Most of clade. Studies of natatanuran phylogeny incorporating these studies also identiA ed the ex-leptodactylid genera both mitochondrial and nuclear genes agreed on the

HHedges.indbedges.indb 361361 11/28/2009/28/2009 11:28:32:28:32 PPMM 362 THE TIMETREE OF LIFE

basal divergence of several African lineages and corrob- around the Cretaceous–Paleogene (K-P) boundary (65.5 orated morphological evidence for a close relationship Ma). Zhang et al. (19), based on multidivtime analyses between Mantellidae and Rhacophoridae (3, 28–30). of mitogenomic data in combination with a single exter- Recent analyses of Microhyloidea using similar multi- nal calibration point, recovered noticeably older age gene data sets have demonstrated the non-monophyly of estimates for several nodes, including a – at least A ve out of nine traditionally recognized subfam- (299 Ma) origin of living anurans and a mid- ilies (2). A consensus for early microhyloid relationships Cretaceous (99.6 Ma) age for the nobleobatrachian clade. is yet unavailable, but at least three studies have provided However, because mitochondrial genes evolve much evidence for a close relationship of Asian Microhylidae faster than most nuclear genes used in other studies, it is and Madagascan Dyscophidae (18, 30, 31). likely that they pose increased risks of mutational satur- During the past few years, molecular divergence time ation and biases in branch length estimation. analyses have resulted in increasingly comprehensive 7 e rise of living anurans shows strong overlap with timetrees for Anura (Table 1, Fig. 2). 7 e earliest stud- major shiJ s in vertebrate faunal compositions in the ies incorporated 7 orne et al.’s (32) relaxed molecular late Permian and Triassic. Both the end Permian mass clock model (divtime) to date primary divergences in and the Triassic extinction episodes repre- Natatanura, suggesting mid- to diver- sented severe losses of amphibian diversity (36) and in siA cation of this clade (33). Subsequent studies imple- parallel to amniote groups, anurans may have taken menting diB erent relaxed-clock models, calibration opportunistic advantage of ecological niche vacancy in points, and sampling strategies have corroborated the redeveloping and increasingly complex vertebrate the late Cretaceous radiation of both Natatanura and ecosystems. Molecular divergence time estimates of all Microhyloidea (29–31, 34). Similar application of 7 orne studies also imply that the major archaeobatrachian lin- and Kishino’s (35) upgraded model, adapted to accom- eages were present on Pangaea before its Jurassic north– modate rate variation across multiple loci (Multidivtime), south breakup into Laurasia and (15, 16, 18). resulted in late Jurassic– time estimates 7 e subsequent formation of distinct anuran faunas for the basal divergences of Neobatrachia (based on A ve in both landmasses is illustrated by three independent genes and A ve calibration points) (21) and Triassic esti- divergences between Laurasian and Gondwanan taxon mates for basal anuran divergences (based on A ve genes pairs: (i) Ascaphidae of North America vs. - and seven calibration points) (16). tidae of , (ii) Rhinophrynidae of North– San Mauro et al. (15) constructed the A rst family-level Central America vs. Pipidae of –, timetree for amphibians using Multidivtime analyses and (iii) Anomocoela of North America–Eurasia vs. of RAG1 sequences and nine calibration points. 7 eir Neobatrachia, originally a Gondwanan group. 7 ese analyses were based on broad phylogenetic sampling of results reinforce the predicted importance of Pangaea frogs and provided conA dence intervals for the age of the breakup in shaping distinct amphibian faunas in both major anuran clades. An expanded study using penal- hemispheres (13). ized likelihood analyses of 84 anuran RAG1 sequences 7 e late Jurassic or early Cretaceous divergences of the and 11 calibration points (26) produced overall younger four major neobatrachian lineages (Fig. 2) constitute a time estimates (Table 1) when the Caudata–Anura split second distinct wave of anuran radiation. Two of the four was arbitrarily A xed at 300 Ma. A parallel study, apply- major lineages are now only represented by few species ing both Multidivtime and penalized likelihood ana- endemic to restricted geographic regions on diB erent lyses using 24 calibration points on a A ve-gene data set ex-Gondwanan landmasses. Heleophrynidae (167 Ma) including 120 anuran taxa (18), resulted in divergence consist of six extant species that occur in the mountain time estimates that were very similar to those of San ranges of the Cape and Transvaal regions of Repulic of Mauro et al. (15), with strong overlap of 95% credibil- South Africa; Sooglossoidea (159 Ma) consists of only ity intervals. 7 ese studies indicate that the evolutionary A ve described species, one of which (Nasikabatrachidae) rise of anuran diversity was a highly episodic process, is endemic to the Indian Western Ghats, and four of with the establishment of archaeobatrachian clades in which (Sooglossidae) occur only on the Seychelles. 7 e the Triassic–early Jurassic (251–199.6 Ma), of the pri- deep split between both families (101 Ma) identiA es mary neobatrachian lineages in the late Jurassic–early each of them as relict lineages that testify for a mid- Cretaceous, of the natatanuran and microhyloid radi- Cretaceous biogeographic link between the Seychelles ations in the late Cretaceous, and of Nobleobatrachia and the Indian subcontinent. A similar deep-time

HHedges.indbedges.indb 362362 11/28/2009/28/2009 11:28:32:28:32 PPMM Eukaryota; Metazoa; Vertebrata; Lissamphibia; Anura 363

intercontinental link is represented by the Cretaceous Mantellidae, Rhacophoridae, Ranidae, Asterophryidae, divergence of the Chilean Calyptocephalellidae and the and Microhylidae) (18, 29). Australo-Papuan Myobatrachoidea (129 Ma). 7 is sug- 7 e timetree that is emerging from the rapid suc- gests that Australobatrachia once had a trans-Gond- cession of molecular analyses provides an increasingly wanan distribution, ranging from South America over detailed temporal framework for anuran evolution. Antarctica to . Besides shedding light on historical biogeography, this Although the nobleobatrachian radiation produced framework allows us to study patterns and rates of evo- approximately half of the currently living anuran spe- lutionary change in morphological, ontogenetic, and cies, molecular divergence time estimates suggest that genomic data. A remaining challenge for future phylo- its initial diversiA cation commenced relatively late, near genetic studies is represented by the explosive radiations the K-P boundary (15, 18). 7 e rapid establishment of Natatanura, Microhyloidea, and Nobleobatrachia, of a large number of lineages that currently represent and several of their families. Resolving these radiations a broad range of ecomorphs (including toads, litter will most likely require alternative (and more expensive) frogs, glass frogs, poison arrow frogs, fossorial frogs, strategies, provided by the expanding A eld of phylog- and several lineages of tree frogs) A ts the pattern of enomics (e.g., using SINES or EST data). 7 e credibility opportunistic radiation in the aJ ermath of the K-P of the anuran timetree is reinforced by the relative con- extinction episode. Given the neotropical distribution sistency among independent studies, despite the use of of most nobleobatrachian families, this radiation is diB erent data sets, calibration points, and methods. In likely to have taken place primarily in South America. addition, although molecular time estimates for some At least four lineages dispersed to other continents anuran nodes are notably older than those derived from in the Tertiary: one lineage of Eleutherodactylidae the fossil record, there are no major incompatibilities reached North America (37, 38), Hylidae and Bufonidae between the two types of data (41). Rather, molecular and attained widespread distributions, probably by disper- fossil analyses can be considered complementary tools to sing through North America and Eurasia ( 39, 40), understand the paleobiological processes and events that and the occurrence of Pelodryadidae in the Australo- shaped the present-day anuran diversity. Papuan realm provides evidence for a Tertiary trans- Antarctic range extension (15). 7 e prevalence of continent-scale endemism in Acknowledgment Ranoides, that is, the clear historical association of fam- Support was provided by the Fonds voor Wetenschap- ilies with a single Gondwanan landmass, suggests that pelijk Onderzoek, Vlaanderen en de Vrije Universiteit continental breakup has played a key role in the distri- Brussel. bution of these frogs (29, 30). 7 e late Cretaceous radia- tions of Natatanura and Microhyloidea indicate that dispersal between Gondwanan landmasses took place at References least until the end Cretaceous (30, 31). In addition, com- 1. AmphibiaWeb, Information on Amphibian Biology parable temporal and spatial divergence patterns within and Conservation, http://amphibiaweb.org/ (Berkeley, the microhyloid and natatanuran radiations suggest that California, 2008). the isolation of their daughter lineages on diB erent con- 2. W. E. Duellman, L. Trueb, Biology of Amphibians (7 e tinents were determined by the same geological events. Johns Hopkins University Press, Baltimore and London, 7 e disruption of terrestrial passages that persisted 1994). between continents long aJ er they started driJ ing apart 3. D. R. Frost et al., Bull. Am. Mus. Nat. Hist. 297, 1 (2006). may have resulted in parallel instances of vicariance in 4. J. C. Avise, G. C. Johns, Proc. Natl. Acad. Sci. U.S.A. 96, both clades (30). 7358 (1999). 7 e late Cretaceous diversiA cation of Natatanura 5. J. D. Lynch, in Evolutionary Biology of the Anurans: Contemporary Research on Major Problems, J. L. Vial, and Microhyloidea imply a survival of multiple lineages Ed. (University of Missouri Press, Columbia, Missouri, across the K-P boundary. Some of the surviving lineages 1973), pp. 133–182. are represented by only few relict species (e.g., Phryno- 6. L. S. Ford, D. C. Cannatella, Herpetol. Monogr. 7, 94 (1993). meridae in Africa, Melanobatrachidae on the Indian 7. A. Haas, J. Zool. Syst. Evol. Res. 35, 179 (1997). subcontinent), but several of their largest families radi- 8. M. Maglia, L. A. Pugener, L. Trueb, Am. Zool. 41, 538 ated substantially in the Paleogene (e.g., Dicroglossidae, (2001).

HHedges.indbedges.indb 363363 11/28/2009/28/2009 11:28:32:28:32 PPMM 364 THE TIMETREE OF LIFE

9. A. Haas, Cladistics 19, 23 (2003). 25. T. Grant et al., Bull. Am. Mus. Nat. Hist. 299, 1 (2006). 10. L. A. Pugener, A. M. Maglia, L. Trueb, Zool. J. Linn. Soc. 26. J. J. Wiens, Am. Nat. 170, S86 (2007). 139, 129 (2003). 27. A. van der Meijden, M. Vences, A. Meyer, Proc. Roy. Soc. 11. S. B. Hedges, L. R. Maxson, Herpetol. Monogr. 7, 27 (1993). Lond. B 271, S378 (2004). 12. J. M. Hay, I. Ruvinsky, S. B. Hedges, L. R. Maxson, Mol. 28. A. van der Meijden, M. Vences, S. Hoegg, A. Meyer, Mol. Biol. Evol. 12, 928 (1995). Phylogenet. Evol. 37, 674 (2005). 13. E. Feller, S. B. Hedges, Mol. Phylogenet. Evol. 9, 509 29. F. Bossuyt, R. M. Brown, D. H. Hillis, D. C. Cannatella, (1998). M. C. Milinkovitch, Syst. Biol. 55, 579 (2006). 14. S. Hoegg, M. Vences, H. Brinkmann, A. Meyer, Mol. Biol. 30. I. Van Bocxlaer, K. Roelants, S. D. Biju, J. Nagaraju, Evol. 21, 1188 (2004). F. Bossuyt, PLoS ONE 1, e74 (2006). 15. D. S. San Mauro, M. Vences, M. Alcobendas, R. Zardoya, 31. A. van der Meijden et al., Mol. Phylogenet. Evol. 44, 1017 A. Meyer, Am. Nat. 165, 590 (2005). (2007). 16. K. Roelants, F. Bossuyt, Syst. Biol. 54, 111 (2005). 32. L. 7 orne, H. Kishino, I. S. Painter, Mol. Biol. Evol. 15, 17. C. Gissi, D. San Mauro, G. Pesole, R. Zardoya,Gene 366, 1647 (1998). 228 (2006). 33. F. Bossuyt, M. C. Milinkovitch, Science 292, 93 (2001). 18. K. Roelants et al., Proc. Natl. Acad. Sci. U.S.A. 104, 887 34. M. Vences et al., Proc. Biol. Sci. 270, 2435 (2003). (2007). 35. L. 7 orne, H. Kishino, Syst. Biol. 51, 689 (2002). 19. P. Zhang, H. Zhou, Y. Q. Chen, Y. F. Liu, L. H. Qu, Syst. 36. J. Benton, e Fossil Record 2 (Chapman & Hall, London, Biol. 54, 391 (2005). 1993), pp. 845. 20. I. Ruvinsky, L. R. Maxson, Mol. Phylogenet. Evol. 5, 533 37. M. P. Heinicke, W. E. Duellman, S. B. Hedges, Proc. Natl. (1996). Acad. Sci. U.S.A. 104, 10092 (2007). 21. S. D. Biju, F. Bossuyt, Nature 425, 711 (2003). 38. S. B. Hedges, W. E. Duellman, M. P. Heinicke, Zootaxa 22. S. K. Dutta, K. Vasudevan, M. S. Chaitra, K. Shankar, 1737, 1 (2008). R. K. Aggarwal, Curr. Sci. 86, 211 (2004). 39. S. A. Smith, P. R. Stephens, J. J. Wiens, Evolution 59, 2433 23. C. R. Darst, D. C. Cannatella, Mol. Phylogenet. Evol. 31, (2005). 462 (2004). 40. J. B. Pramuk, T. Robertson, J. W. Sites, Jr., B. P. Noonan, 24. J. J. Wiens, J. W. Fetzner, C. L. Parkinson, T. W. Reeder, Global Ecol. Biogeogr. 17, 72 (2008). Syst. Biol. 54, 719 (2005). 41. D. Marjanovic, M. Laurin, Syst. Biol. 56, 369 (2007).

HHedges.indbedges.indb 364364 11/28/2009/28/2009 11:28:33:28:33 PPMM

Top View